The offeror will employ two different strains of mice which have been genetically engineered to be highly susceptible to colon tumor. One strain of mice will have a deletion in the APC gene, resulting in a molecular phenotype similar to human FAP. The second strain of mice of mice will have a deletion in the MSH-2 gene, resulting in a molecular pathology similar to a subset of human HNPCC. A diet comparable to that routinely eaten in the developed world, relatively high in fat and relatively low in calcium, will be employed to modulate the tumor incidence and multiplicity in these genotypically altered mice. Samples from these mice will be taken at various time points either from control animals or animals treated with curcumin, rutin, sulindac or high Calcium high vitamin D. At these various times either colonic mucosa or specific lesions will be examined for a variety of endpoints including DNA ploidy, Proliferation status, nuclear morphometry [size, shape, texture], nucleolar morphometry, cytokeratins and adhesion molecules. In the case of the colonic mucosae , prior to appearance of large discrete lesions, analysis of individual crypts for these various endpoints should be employed. Increased or decreased expression of these various markers will be compared with the final chemopreventive activity of curcumin, rutin, sulindac and high calcium to determine whether modulation of specific chemical markers is associated with long term chemopreventive activity. In addition the studies will yield the therapeutic efficacy of these high priority preventive agents in these models which closely approximate their human counterpart.